Polar codes have been proved as a strong benchmark when it comes to channel polarization of the next generation of mobile communication, 5G. Introduced by Arikan, polar codes are the first type of channel codes in which their capacity can be achieved by means of theoretical proof, assuming an infinite code length and successive-cancellation decoder (SC). However, finite-length codes have a lower performance in terms of error rate under the SC decoder, in addition to presenting a higher latency due to the sequential nature of the decoder, in which the information bits are decoded serially.Considering that the SC decoder for finite length codes is sub-optimal, Tal and Vardy proposed a listbased Successive Cancellation Decoder (SCL) in which it approaches the maximum likelihood (ML) performance for a list size large enough at the cost of greater complexity. Furthermore, Cyclic Redundancy Check Codes (CRC) can be effortlessly implemented in order to improve the SCL algorithm by increasing its minimum distance. This combination makes polar codes a powerful decoding scheme, despite their high complexity and a serial inherent nature. Therefore, several variants of the decoder have been proposed in order to reduce the computational complexity.As an alternative to the serial nature of the SC decoder, Arikan proposed an iterative decoder algorithm with great potential for parallelism based on Belief Propagation (BP) under the codification diagram of the polar codes. Nevertheless, despite the algorithm outperforming the SC decoder, being a strong candidate in applications which demand a high data rate and more suitable to hardware implementations, your performance is not comparable to the CRC-aided SCL decoder. Therefore, several efforts have been made to enhance the decoder performance. It has been shown that finite-length polar codes under BP decoding can be improved when semi-polarized channels are additionally protected by check nodes, or with an augmented polar code. Unfortunately, these approaches have inferior performance when compared to SCL decoding. Besides, they require an adjusted code structure and are thus not compatible with the standardized polar code in the literature. Therefore, our project consists of developing a new BP decoder, based on Q-Learning, which maintains the polar code structure, and obtains a performance consistently superior to the standard BP and SC decoders.
scite is a Brooklyn-based organization that helps researchers better discover and understand research articles through Smart Citations–citations that display the context of the citation and describe whether the article provides supporting or contrasting evidence. scite is used by students and researchers from around the world and is funded in part by the National Science Foundation and the National Institute on Drug Abuse of the National Institutes of Health.
hi@scite.ai
10624 S. Eastern Ave., Ste. A-614
Henderson, NV 89052, USA
Copyright © 2024 scite LLC. All rights reserved.
Made with 💙 for researchers
Part of the Research Solutions Family.